80/20 is the new Half-Ass
swyx
Posted on April 28, 2021
The Pareto Principle is making you lazy.
Let me be more precise: The Pareto distribution is a useful model of power law effects in real life. But people are using it poorly, primarily as an excuse to be lazy.
This thought was triggered by Shaan Puri's newsletter featuring Steph Smith this week (both of whom I greatly respect, this is a jumping off point, not a dunk):
You hear 80/20 rules a lot in premium mediocre circles. If you want to signal that you are smarter than the average 80/20 bear, you might refer to the "high order bit" instead. Same shit, different status signal.
The spirit of the idea is sound. It's great for 280 character tweets and 5 minute soundbites. But idk if it's good for people who finish things. At best, I don't think it's sufficient for execution. At worst, it's just intellectually dishonest.
Look at reactions like this:
"Love this [80/20] framing! Great way to remove the fluff and get to the core".
I'm sorry but the remaining 80% is not "the fluff":
- People forget that the devil is in the details. The first 20% everyone knows to say on Twitter. The remaining 80% is the ugly, nasty, hacky, unglamorous shit nobody talks about unless you've got time to sweat the details (unless you've already moved on to the next 80/20?).
- The more popular the 80/20 meme becomes, the less competition you will have as someone who knows how to take things to the finish line.
- People forget that causal attribution is subject to narrative fallacy. Ask a successful person what their 80/20 was and they'll confidently tell you in hindsight. The truth is at the time they had 4 other bets also going on that just didn't work out.
- The popular saying in medical school is that "50% of what you learn will be wrong — we just don't know which 50%". And that is in goddamn medical science with double blind randomized clinical trials. We don't have 1% of that rigor in popular anecdata which you base your 80/20 inferences on.
- People forget that distributions aren't always Pareto. When you assert that "an 80/20" exists, you are asserting useful dimensionality reduction. Sometimes complex, or even linear, things just do not have an "80/20".
- The FAA's Parachute Rigger Handbook has 350 pages of densely packed advice. Please do not try to 80/20 your parachute packing.
I'm reminded of that classic movie Click, where Adam Sandler finds a magic remote that lets him "80/20 his life":
Don't spend your life spraying 20% effort all over the place, hoping for 80% results, only to look back and wonder why you never hit 100% on anything.
Edit: The always excellent David Golden wrote a wonderful response to this piece highlighting the cases where 80/20 is actually very justified, and I agree!
This topic is loosely related to Epistemology, which I've written about briefly.
Posted on April 28, 2021
Join Our Newsletter. No Spam, Only the good stuff.
Sign up to receive the latest update from our blog.